Import Libraries¶
In [1]:
import numpy as np
import pandas as pd
from pandas import DataFrame as df
import matplotlib.pyplot as plt
import seaborn as sns
Load Dataset¶
In [2]:
canada = pd.read_excel('https://s3-api.us-geo.objectstorage.softlayer.net/cf-courses-data/CognitiveClass/DV0101EN/labs/Data_Files/Canada.xlsx',
sheet_name='Canada by Citizenship',
skiprows=range(20),
skipfooter=2)
In [3]:
canada.head()
#spare df
canada1 = canada
Rename columns¶
In [4]:
canada.rename(columns={'OdName':'Country', 'AreaName':'Continent', 'RegName':'Continent-Region'}, inplace=True)
canada.drop(['AREA','REG','DEV','Type','Coverage'], inplace=True, axis='columns')
canada.isnull().sum().sum()
Out[4]:
In [5]:
canada.head(5)
Out[5]:
In [6]:
canada.index.values
Out[6]:
Reassign Index¶
In [7]:
canada.set_index('Country', inplace=True)
In [8]:
canada.head()
Out[8]:
Convert columns to str¶
In [9]:
canada.dtypes
canada.columns = list(map(str,canada.columns));
Subsetting¶
In [10]:
canada[canada['Continent']=='Asia'].head(4)
Out[10]:
Viewing Line chart¶
In [11]:
years = list(map(str,range(1980,2014)))
india_imgt = canada.loc['India', years]
india_imgt.plot(figsize=(10,6));
india_imgt.plot(kind='line');
plt.title('Immigration from India');
plt.ylabel('Number of immigrants');
plt.xlabel('Years');
Plotting two countries and comparing¶
In [12]:
canada.loc[['China','India'], years].transpose().plot(figsize =(10,7));
plt.title('Immigration of China and India');
plt.xlabel('Years');
plt.ylabel('Immigrants');
Area Plots¶
Unstacked¶
In [13]:
top5_clean = canada.head(5)[years].transpose()
top5_clean.index
top5_clean.index = top5_clean.index.map(int)
top5_clean.plot(kind="area", stacked=False, figsize=(20,10), alpha=0.25)
plt.title('Immigration trends in top five countries (1980-2013)')
plt.xlabel('Years')
plt.ylabel('Immigrants count')
plt.show()
Stacked¶
In [14]:
top5_clean.plot(kind="area", stacked=True, figsize=(20,10), alpha=0.25)
plt.title('Immigration trends in top five countries (1980-2013)')
plt.xlabel('Years')
plt.ylabel('Immigrants count')
plt.show()
Bar PLots and Histograms¶
In [15]:
df_can = canada
Bar¶
In [16]:
df_can['2000'].plot(kind='hist', figsize=(20,10))
plt.title('Immigration trends in 195 countries in 2000')
plt.xlabel('Number of Immigrants')
plt.ylabel('Number of Countries')
plt.show()
Histogram¶
In [17]:
df_can.loc[['India','China','Denmark','Norway','France','Germany'], years].transpose().plot.hist()
plt.title('Immigration trends in 6 countries in 1980-2013')
plt.xlabel('Number of Immigrants')
plt.ylabel('Number of Countries')
plt.show()
Vertical Bar Plots¶
In [18]:
india = df_can.loc['India', years]
india.plot(kind='bar', figsize=(20,10))
plt.title('Immigration trends India immigrants from 1980 to 2013')
plt.xlabel('Years')
plt.ylabel('India immigrants from 1980 to 2013')
plt.show()
Horizontal Bar Plots¶
In [19]:
india = df_can.loc['India', years]
india.plot(kind='barh', figsize=(20,10))
plt.title('Immigration trends India immigrants from 1980 to 2013')
plt.xlabel('India immigrants from 1980 to 2013')
plt.ylabel('Years')
plt.show()
Using annotate¶
In [20]:
df_india = df_can.loc['India',years]
df_india.plot(kind='bar', figsize=(10,6))
plt.title("Indian immigration to Canada from 1980 to 2013")
plt.xlabel("Year")
plt.ylabel("Number of immigrants")
plt.annotate('', #arrow title
xy=(21,35000), #x,y of arrow head
xytext=(7,15000), #x, y of arrow tail
xycoords='data', #keep unchanged
arrowprops=dict(arrowstyle='->',color='red') #arrow style with color
)
plt.annotate('IT Boom', #add text to arrow
xy=(13,28000), #x, y of text position
rotation=40, # counter clockwise rotate text by angle
xycoords='data', #keep unchanged
va='top', #position text
ha='left') #position text
plt.show()
Lab 4 starts here ---¶
Scatter¶
In [21]:
indopak = canada.loc[['India','Pakistan'], years].transpose()
In [22]:
indopak.reset_index(inplace=True)
indopak.rename(columns={'Country':'index','index':'Year'}, inplace=True)
indopak.head()
Out[22]:
In [23]:
import numpy as np
x = indopak['Year'].astype('int64')
y = indopak['India']
fit = np.polyfit(x,y,deg=1)
print(fit)
indopak.plot(kind='scatter', x='Year', y='India', color='blue', figsize=(10,6))
plt.title('Total immigrants from India 1980-2013')
plt.xlabel("years")
plt.ylabel("Number of immigrants")
#plt.plot(x, fit[0]*x+fit[1], color='red')
plt.show();
Plotting 2 Scatter Plots in the Same Figure¶
In [24]:
ax0 = indopak.plot(kind='scatter', x='Year', y='India', color='blue', figsize=(10,6))
indopak.plot(kind='scatter', x='Year', y='Pakistan', color='green', figsize=(10,6), ax=ax0)
plt.title('Total immigrants from India and Pakistan 1980-2013')
plt.xlabel("years")
plt.ylabel("Number of immigrants")
ax0.legend(['India','Pakistan'], loc='upper left')
plt.show();
Plotting 2 Scatter Plots in the Same Figure¶
In [25]:
norm_india = (indopak['India'] - indopak['India'].min()) / (indopak['India'].max() - indopak['India'].min())
norm_pak = (indopak['Pakistan'] - indopak['Pakistan'].min()) / (indopak['Pakistan'].max() - indopak['Pakistan'].min())
ax0 = indopak.plot(kind='scatter', x='Year', y='India', color='blue', s=norm_india*200, figsize=(10,6))
indopak.plot(kind='scatter', x='Year', y='Pakistan', color='green', s=norm_india*200, figsize=(10,6), ax=ax0)
plt.title('Total immigrants from India and Pakistan 1980-2013')
plt.xlabel("years")
plt.ylabel("Number of immigrants")
ax0.legend(['India','Pakistan'], loc='upper left')
plt.show();
Waffle Charts¶
Exp 4 Assignment included¶
In [26]:
%matplotlib inline
import matplotlib as mpl
import matplotlib.pyplot as plt
import matplotlib.patches as mpatches
In [27]:
df_can['Total'] = df_can.sum (axis = 1)
In [28]:
# let's create a new dataframe for these three countries
df_dsn = df_can.loc[['Denmark', 'Norway', 'Sweden'], :]
# let's take a look at our dataframe
df_dsn
Out[28]:
In [29]:
# compute the proportion of each category with respect to the total
total_values = sum(df_dsn['Total'])
category_proportions = [(float(value) / total_values) for value in df_dsn['Total']]
# print out proportions
for i, proportion in enumerate(category_proportions):
print (df_dsn.index.values[i] + ': ' + str(proportion))
In [30]:
width = 40 # width of chart
height = 10 # height of chart
total_num_tiles = width * height # total number of tiles
print ('Total number of tiles is ', total_num_tiles)
In [31]:
# compute the number of tiles for each catagory
tiles_per_category = [round(proportion * total_num_tiles) for proportion in category_proportions]
# print out number of tiles per category
for i, tiles in enumerate(tiles_per_category):
print (df_dsn.index.values[i] + ': ' + str(tiles))
In [32]:
# initialize the waffle chart as an empty matrix
waffle_chart = np.zeros((height, width))
# define indices to loop through waffle chart
category_index = 0
tile_index = 0
# populate the waffle chart
for col in range(width):
for row in range(height):
tile_index += 1
# if the number of tiles populated for the current category is equal to its corresponding allocated tiles...
if tile_index > sum(tiles_per_category[0:category_index]):
# ...proceed to the next category
category_index += 1
# set the class value to an integer, which increases with class
waffle_chart[row, col] = category_index
print ('Waffle chart populated!')
In [33]:
waffle_chart
Out[33]:
In [34]:
# instantiate a new figure object
fig = plt.figure()
# use matshow to display the waffle chart
colormap = plt.cm.coolwarm
plt.matshow(waffle_chart, cmap=colormap)
plt.colorbar()
Out[34]:
In [35]:
# instantiate a new figure object
fig = plt.figure()
# use matshow to display the waffle chart
colormap = plt.cm.coolwarm
plt.matshow(waffle_chart, cmap=colormap)
plt.colorbar()
# get the axis
ax = plt.gca()
# set minor ticks
ax.set_xticks(np.arange(-.5, (width), 1), minor=True)
ax.set_yticks(np.arange(-.5, (height), 1), minor=True)
# add gridlines based on minor ticks
ax.grid(which='minor', color='w', linestyle='-', linewidth=2)
plt.xticks([])
plt.yticks([])
# compute cumulative sum of individual categories to match color schemes between chart and legend
values_cumsum = np.cumsum(df_dsn['Total'])
total_values = values_cumsum[len(values_cumsum) - 1]
# create legend
legend_handles = []
for i, category in enumerate(df_dsn.index.values):
label_str = category + ' (' + str(df_dsn['Total'][i]) + ')'
color_val = colormap(float(values_cumsum[i])/total_values)
legend_handles.append(mpatches.Patch(color=color_val, label=label_str))
# add legend to chart
plt.legend(handles=legend_handles,
loc='lower center',
ncol=len(df_dsn.index.values),
bbox_to_anchor=(0., -0.2, 0.95, .1)
)
Out[35]:
Or use a waffle function¶
In [36]:
def create_waffle_chart(categories, values, height, width, colormap, value_sign=''):
# compute the proportion of each category with respect to the total
total_values = sum(values)
category_proportions = [(float(value) / total_values) for value in values]
# compute the total number of tiles
total_num_tiles = width * height # total number of tiles
print ('Total number of tiles is', total_num_tiles)
# compute the number of tiles for each catagory
tiles_per_category = [round(proportion * total_num_tiles) for proportion in category_proportions]
# print out number of tiles per category
for i, tiles in enumerate(tiles_per_category):
print (df_dsn.index.values[i] + ': ' + str(tiles))
# initialize the waffle chart as an empty matrix
waffle_chart = np.zeros((height, width))
# define indices to loop through waffle chart
category_index = 0
tile_index = 0
# populate the waffle chart
for col in range(width):
for row in range(height):
tile_index += 1
# if the number of tiles populated for the current category
# is equal to its corresponding allocated tiles...
if tile_index > sum(tiles_per_category[0:category_index]):
# ...proceed to the next category
category_index += 1
# set the class value to an integer, which increases with class
waffle_chart[row, col] = category_index
# instantiate a new figure object
fig = plt.figure()
# use matshow to display the waffle chart
colormap = plt.cm.coolwarm
plt.matshow(waffle_chart, cmap=colormap)
plt.colorbar()
# get the axis
ax = plt.gca()
# set minor ticks
ax.set_xticks(np.arange(-.5, (width), 1), minor=True)
ax.set_yticks(np.arange(-.5, (height), 1), minor=True)
# add dridlines based on minor ticks
ax.grid(which='minor', color='w', linestyle='-', linewidth=2)
plt.xticks([])
plt.yticks([])
# compute cumulative sum of individual categories to match color schemes between chart and legend
values_cumsum = np.cumsum(values)
total_values = values_cumsum[len(values_cumsum) - 1]
# create legend
legend_handles = []
for i, category in enumerate(categories):
if value_sign == '%':
label_str = category + ' (' + str(values[i]) + value_sign + ')'
else:
label_str = category + ' (' + value_sign + str(values[i]) + ')'
color_val = colormap(float(values_cumsum[i])/total_values)
legend_handles.append(mpatches.Patch(color=color_val, label=label_str))
# add legend to chart
plt.legend(
handles=legend_handles,
loc='lower center',
ncol=len(categories),
bbox_to_anchor=(0., -0.2, 0.95, .1)
)
In [37]:
width = 40 # width of chart
height = 10 # height of chart
categories = df_dsn.index.values # categories
values = df_dsn['Total'] # correponding values of categories
colormap = plt.cm.coolwarm # color map class
In [38]:
create_waffle_chart(categories, values, height, width, colormap)
In [39]:
indpakchi = canada.loc[['India', 'Pakistan', 'China'], :]
categories = list(indpakchi.index.values)
values = indpakchi['Total']
create_waffle_chart(categories, values, height, width, colormap)
Total Count of Immigration across all countries¶
In [40]:
canada['Total'].plot(kind='pie',figsize=(10,10));
Plotting the total immigration data by each continent¶
In [41]:
can_continent = canada.groupby('Continent', axis='index').sum()
can_continent.head()
Out[41]:
In [42]:
can_continent['Total'].plot(kind='pie', figsize=(7, 7));
Modifying the graph¶
In [43]:
can_continent['Total'].plot(kind='pie', figsize=(7, 7),
startangle = 180, autopct = '%1.1f%%')
plt.title('Immigration To Canada By Continents Between 1980 - 2013')
plt.axis('equal')
plt.legend(labels = can_continent.columns, loc='upper right')
plt.show()
Plotting Immigration Data by Development Status¶
In [44]:
can_devstatus = canada.groupby('DevName', axis='index').sum()
can_devstatus.head()
Out[44]:
In [45]:
explodeList = [0.3,0]
colorList=['blue','magenta']
can_devstatus['Total'].plot(kind='pie', figsize=(7,7), startangle=90,autopct='%1.1f%%',
shadow=True, labels = None, colors = colorList, explode = explodeList)
plt.title('Immigration from 1980-2013 by development status')
plt.axis('equal')
plt.legend(labels= can_devstatus.index,loc ='upper right')
plt.show()
Plotting a pie chart for Immigration with respect to Country-Region¶
In [46]:
can_contreg = canada.groupby('Continent-Region', axis='index').sum()
can_contreg.head()
Out[46]:
In [47]:
explodeList = [0.1]*22
#colorList=['blue','magenta']
can_contreg['Total'].plot(kind='pie', figsize=(10,10), startangle=90,autopct='%1.1f%%',
shadow=True, labels = None, explode = explodeList)
plt.title('Immigration from 1980-2013 by Continent-Region')
plt.axis('equal')
plt.legend(labels= can_contreg.index,loc ='upper right')
plt.show()
Box Plots¶
Plotting a boxplot for Immigration from India across the years¶
In [48]:
india = canada.loc['India', years]
india.plot(kind='box', figsize=(10,6))
plt.title('Box plot of India from 1980-2013')
plt.ylabel('Number of immigrants')
plt.show()
Comparative Boxplot for Immigration from India and Pakistan¶
In [49]:
indopak = canada.loc[['India','Pakistan'], years].transpose()
indopak.plot(kind='box', figsize=(10,6))
plt.title('Box plot of India and Pakistan from 1980-2013')
plt.ylabel('Number of immigrants')
plt.show()
Describing the Ind-Pak Immigration Data¶
In [50]:
indopak.describe()
Out[50]:
Modifying the Boxplot for Ind-Pak Data¶
In [51]:
indopak.plot(kind='box', figsize=(10,6), color='red', vert=False)
plt.title('Box plot of India and Pakistan from 1980-2013')
plt.xlabel('Number of immigrants')
plt.show()
Creating 4 subplots of for boxplots of top 5 countries In Europe, Asia, Africa and Oceania¶
In [52]:
asia = canada[(canada['Continent']=='Asia')].sort_values('Total', ascending=False).head()
eur = canada[(canada['Continent']=='Europe')].sort_values('Total', ascending=False).head()
afr = canada[(canada['Continent']=='Africa')].sort_values('Total', ascending=False).head()
ocn = canada[(canada['Continent']=='Oceania')].sort_values('Total', ascending=False).head()
asia = asia[years].transpose()
eur = eur[years].transpose()
afr = afr[years].transpose()
ocn = ocn[years].transpose()
PLots¶
In [53]:
fig = plt.figure();
ax1 = fig.add_subplot(1,2,1);
eur.plot(kind='box', color='blue', vert=False, ax=ax1);
ax2 = fig.add_subplot(1,2,2);
asia.plot(kind='box', color='blue', vert=False, ax=ax2);
fig2 = plt.figure();
ax3 = fig2.add_subplot(1,2,1);
afr.plot(kind='box', color='blue', vert=False, ax=ax3);
ax4 = fig2.add_subplot(1,2,2);
ocn.plot(kind='box', color='blue', vert=False, ax=ax4);
Outliers¶
In [54]:
temp = asia.describe()
Q1 = temp.loc['25%', 'Philippines']
Q3 = temp.loc['75%', 'Philippines']
IQR = Q3-Q1
Outlier1 = Q3 + 1.5*IQR
Outlier2 = Q1 - 1.5*IQR
print(Outlier1)
print(Outlier2)
asia[asia['Philippines']>Outlier1]
Out[54]:
Word Clouds¶
importing¶
In [55]:
from wordcloud import WordCloud, STOPWORDS
In [56]:
alice_novel = open('/Users/home/Downloads/1462444-088c12e7f74fcff9be6c4faa556f961cb7a675bd/alice.txt', 'r').read()
In [57]:
stopwordsremove = set(STOPWORDS)
alice_wc = WordCloud(background_color = 'white', max_words = 2000,
stopwords = stopwordsremove)
In [58]:
alice_wc.generate(alice_novel)
plt.figure(figsize = (10,15))
plt.imshow(alice_wc)
plt.show()
In [59]:
stopwordsremove = set(STOPWORDS)
stopwordsremove.add('said')
alice_wc = WordCloud(background_color = 'white', max_words = 3000,
stopwords = stopwordsremove)
alice_wc.generate(alice_novel)
fig = plt.figure()
fig.set_figheight(20)
fig.set_figwidth(20)
plt.imshow(alice_wc)
plt.axis('off')
plt.show()
Maps¶
In [60]:
import folium
Map of Dahisar¶
In [61]:
worldmap = folium.Map(location=[19.25, 72.86], zoom_start=15)
worldmap
Out[61]:
In [62]:
worldmap = folium.Map(location=[19.25, 72.86], zoom_start=14, tiles='Stamen Toner')
worldmap
Out[62]:
In [63]:
worldmap = folium.Map(location=[19.25, 72.86], zoom_start=14, tiles='Stamen Terrain')
worldmap
Out[63]:
Loading different dataset¶
In [64]:
incidents = pd.read_csv('https://s3-api.us-geo.objectstorage.softlayer.net/cf-courses-data/CognitiveClass/DV0101EN/labs/Data_Files/Police_Department_Incidents_-_Previous_Year__2016_.csv')
print('Dataset downloaded and read into a pandas dataframe!')
print(incidents.head())
print(incidents.shape)
In [65]:
limit = 100
inc = incidents.iloc[0:limit,:]
In [66]:
latitude = 37.77
longitude = -122.42
map1 = folium.Map(location = [latitude, longitude], zoom_start=12)
map1
Out[66]:
In [67]:
#given information about city coordinates
latitude= 37.77
longitude = -122.42
#create map using folium library and Map class
map1 = folium.Map(location = [latitude, longitude], zoom_start=12)
map1
#incident we will impose on the map
incident = folium.map.FeatureGroup()
#reduce size of dataset
df1 = incidents.iloc[0:limit,:]
#loop through the crimes by their lat Y and long X and create markers
for lat, long in zip(df1.Y, df1.X):
incident.add_child(folium.CircleMarker([lat, long], radius = 5, color = 'yellow', fill=True,
fill_color='blue').add_to(incident))
#add pop ups
lats = list(df1.Y)
longs = list(df1.X)
labels = list(df1.Category)
for lat, long, labs in zip(lats, longs, labels):
folium.Marker([lat, long], popup=labs).add_to(map1)
#add incidents/crimes to the map
map1.add_child(incident)
Out[67]:
In [68]:
#given information about city coordinates
latitude= 37.77
longitude = -122.42
#create map using folium library and Map class
map1 = folium.Map(location = [latitude, longitude], zoom_start=12)
map1
#incident we will impose on the map
incident = folium.map.FeatureGroup()
#reduce size of dataset
df1 = incidents.iloc[0:limit,:]
#loop through the crimes by their lat Y and long X and create markers
lats = list(df1.Y)
longs = list(df1.X)
labels = list(df1.Category)
for lat, long, labs in zip(lats, longs, labels):
folium.CircleMarker([lat, long], popup=labs, radius=5, color='yellow', fill=True,
fill_color='blue').add_to(map1)
#add incidents/crimes to the map
map1
Out[68]:
In [69]:
#given information about city coordinates
latitude= 37.77
longitude = -122.42
#create map using folium library and Map class
map1 = folium.Map(location = [latitude, longitude], zoom_start=12)
map1
#incident we will impose on the map
from folium import plugins
incident = plugins.MarkerCluster().add_to(map1)
#reduce size of dataset
df1 = incidents.iloc[0:limit,:]
#loop through the crimes by their lat Y and long X and create markers
lats = list(df1.Y)
longs = list(df1.X)
labels = list(df1.Category)
for lat, long, labs in zip(lats, longs, labels):
folium.CircleMarker([lat, long], popup=labs, radius=5, color='yellow', fill=True,
fill_color='blue').add_to(incident)
#add incidents/crimes to the map
map1
Out[69]:
Choropeth Markers¶
Assignment 1¶
Which two countries have similar immigration trends over the years 1980-2013?¶
In [70]:
canada['total_immigration'] = canada.sum(axis='columns')
canada.sort_values(by='total_immigration', ascending=False, axis='index', inplace=True)
canada.head(10)
Out[70]:
In [71]:
canada.loc[canada.index.tolist(), years].transpose().plot(figsize=(15,15),legend=False);
plt.title('Immigration of 195 countries');
plt.xlabel('Years');
plt.ylabel('Immigrants');
Plotting three similar countries after year 2009 in the top 10 list.¶
In [72]:
canada.loc[['India','China','Philippines'], years].transpose().plot(figsize=(15,15));
plt.title('Immigration of India, China and Philippines');
plt.xlabel('Years');
plt.ylabel('Immigrants');
Since 2009, India China and Philippines have shown similar trends.¶
As it is impossible to compare 195 countries and their charts, Last 3 year's average is taken and compared in the following steps. The selection of n years is random.¶
Taking avg. of last 3 years and cleaning the Series¶
In [73]:
canada['avglast3'] = (canada['2011']+canada['2012']+canada['2013'])/3
l3 = canada['avglast3']
l3 = l3.sort_values(ascending = False )
l3 = l3.round()
l3 = l3[l3>2]
In [74]:
l3.head(10)
Out[74]:
Removing outliers¶
In [75]:
l5 = l3[l3<11000];
Calculating Standard Deviation for bins¶
In [76]:
np.std(l5)
Out[76]:
Creating bins¶
In [77]:
type(l5)
dl5 = df(l5)
dl5
ahg = pd.Series(range(0,12000,1630))
bins = ahg
dl5['binned'] = pd.cut(dl5['avglast3'], bins)
Plotting the bar graph¶
In [78]:
out = dl5['binned']
ax = out.value_counts(sort=False).plot.bar(rot=0, color="b", figsize=(16,4))
plt.show()
Bin values¶
In [79]:
dl5['binned'].value_counts(sort=True)
Out[79]:
Countries showing similar trends (2011-2013), grouped¶
In [80]:
print('(9780, 11410]')
print(' {ho} {jo}'.format(ho = l5.index[0], jo = l5[0]))
print('')
print('')
print('(8150, 9780]')
print(' {ho} {jo}'.format(ho = l5.index[1], jo = l5[1]))
print('')
print('')
print('(6520, 8150]')
print(' {ho} {jo}'.format(ho = l5.index[2], jo = l5[2]))
print('')
print('')
print('(4890, 6520]')
print(' {jo}'.format(jo = l5[3:7]))
print('')
print('')
print('(3260, 4890]')
print(' {jo}'.format(jo = l5[7:14]))
print('')
print('')
print('(1630, 3260]')
print(' {jo}'.format(jo = l5[14:28]))
print('')
print('')
print('(0, 1630]')
print(' {jo}'.format(jo = l5[28:172]))
Plots¶
Comparison on basis of average of immigrants between years 2011 and 2013¶
Top 3 - excluding India, China and Philippines¶
In [81]:
a1 = l5[0:3].index.tolist()
canada.loc[a1, years].transpose().plot(figsize=(15,15));
plt.title('Immigration of {as1}'.format(as1 = a1));
plt.xlabel('Years');
plt.ylabel('Immigrants');
Countries in Range - (3260, 6520]¶
In [82]:
years = list(map(str,range(1980,2014)))
a2 = l5[4:14].index.tolist()
canada.loc[a2, years].transpose().plot(figsize=(15,12));
plt.title('Immigration of {as2}'.format(as2 = a2));
plt.xlabel('Years');
plt.ylabel('Immigrants');
plt.ylim((0,7000));
Countries in Range - (1630, 3260]¶
In [83]:
years1 = list(map(str,range(2000,2014)))
a4 = l5[14:28].index.tolist()
canada.loc[a4, years1].transpose().plot(figsize=(21,13));
plt.title('Immigration of {as4}'.format(as4 = a4));
plt.xlabel('Years');
plt.ylabel('Immigrants');
plt.ylim((0,6100));
Countries in Range - (1000, 1630]¶
In [84]:
a4 = l5[29:42].index.tolist()
canada.loc[a4, years1].transpose().plot(figsize=(19,13));
plt.title('Immigration of {as4}'.format(as4 = a4));
plt.xlabel('Years');
plt.ylabel('Immigrants');
plt.ylim((400,2800));
Key comparisons and analysis¶
India¶
In [85]:
canada.loc[['India'], years].transpose().plot(figsize =(15,15));
plt.title('Immigration of India');
plt.xlabel('Years');
plt.ylabel('Immigrants');
Inference¶
Fist spike is observed during FY 1985-1986. Next spike is observed in 1990-1991. This can be attributed to Liberalization. Following that in 1995-1996 india became the fifth largest economy in world with 3.9% share in world GDP. We can see that 1999-2000 there is a spike. This can be attributed to the Y2k bug solution.¶
India and China¶
In [86]:
canada.loc[['China','India'], years].transpose().plot(figsize =(15,15));
plt.title('Immigration of China and India');
plt.xlabel('Years');
plt.ylabel('Immigrants');
Lebanon and Sri Lanka¶
In [87]:
a9 = ['Lebanon', 'Sri Lanka']
canada.loc[['Lebanon', 'Sri Lanka'], years].transpose().plot(figsize=(15,15));
plt.title('Immigration of {as9}'.format(as9 = a9));
plt.xlabel('Years');
plt.ylabel('Immigrants');
Inference¶
Sri Lanka¶
Increased privatization, economic reform, and a stress on export-oriented growth helped improve the economic performance, increasing GDP growth to 7% in 1993. This attributes to the drop in immigrants from Sri Lanka.¶
Lebanon¶
The Lebanese Civil war which ended in 1990, severely damaged the Lebanese economy. This dropped continously for four years until 1994, when the Real GDP rose to 8%. This was possible because of the government's 20 billion dollar reconstruction program, which managed to create new jobs, making way for immigrants to return back¶
Pakistan and South Korea (Republic of Korea)¶
In [88]:
a9 = ['Pakistan',' Republic of Korea']
canada.loc[['Pakistan', 'Republic of Korea'], years].transpose().plot(figsize=(15,15));
plt.title('Immigration of {as9}'.format(as9 = a9));
plt.xlabel('Years');
plt.ylabel('Immigrants');
Inference¶
Both countries follow a similar trendline. The graph shows steady increase till about the year 1995. Both countries show a drop in the 2000-2002 period.¶
Pakistan:¶
In 1994 inflation rate in Pakistan went to a whopping 14.5%. The following year the GDP growth rate dropped from 4.5% to 1.7%, lowest in decades. Pakistan's GDP growth rate dropped from 7.68% to 1.7% in a period of 4 years.¶
In 1999 Pervez Musharraf lead a millitary coup and overthrew the Nawaz Sharif government. This resulted in Liberalization. During Musharraf's term (1999-2008), around 11,9 million new jobs were created. Therefore this can be attributed to the drop in immigrants. Also, immigration was affected due to various events that happened in USA during 2001.¶
South Korea:¶
Korea shows a spike in the year 1997, this indicates the Asian Financial Crisis. The Korean Won began to depreciate heavily. Within months, a third of Korea's merchant banks were closed. Korean economy shrinked at an average of -6.65% per year. Conglomerates became a casuality of this.¶
In 2000, the Korean economy shifted to a more market oriented economy. Growth rates touched closer to 11%, highest in decades. This resulted a drop in immigrants.¶
Individual trend analysis¶
Romania¶
In [89]:
years = list(map(str,range(1980,2014)))
canada.loc[['Romania'], years].transpose().plot(figsize=(15,15));
plt.title('Immigration of Romania');
plt.xlabel('Years');
plt.ylabel('Immigrants');
Philippines¶
In [90]:
years = list(map(str,range(1980,2014)))
canada.loc[['Philippines'], years].transpose().plot(figsize=(15,15));
plt.title('Immigration of Philippines');
plt.xlabel('Years');
plt.ylabel('Immigrants');
United Kingdom¶
In [91]:
years = list(map(str,range(1980,2014)))
canada.loc[['United Kingdom of Great Britain and Northern Ireland'], years].transpose().plot(figsize=(15,15));
plt.title('Immigration of United Kingdom of Great Britain and Northern Ireland');
plt.xlabel('Years');
plt.ylabel('Immigrants');
Inference¶
Points of interests: 1984-1986 and 1994¶
In 1984, an all time high unemployment rate of 11.9% was witnessed in the UK, forcing people to search for jobs in other countries. This attributes to the spike in 1984-1985.¶
The John Major government in 1990 fucussed to creating new jobs, which substancially rediced the unemployment rate. This resulted in tyhe immigrants returning back to the UK, which can be seen as the graph is seen dipping post 1994.¶
Viet Nam¶
In [92]:
years = list(map(str,range(1980,2014)))
canada.loc[['Viet Nam'], years].transpose().plot(figsize=(15,15));
plt.title('Immigration of Viet Nam');
plt.xlabel('Years');
plt.ylabel('Immigrants');
Inference¶
Points of interest: 1983-1998 and 1993,¶
Vietnam entered its Third Five-Year-Plan scheme in 1981. This resulted in a massive economic failure, slowing down the economic growth heavily. The government swiftly corrected its path. Increase in exports is the main reason for immigrants to drop post 1993. This was done to bring back the economy on track.¶
Assignment 2¶
Mumbai Crimes (line)¶
In [93]:
indcri = pd.read_csv('https://storage.googleapis.com/kagglesdsdata/datasets%2F1850%2F3879%2Fcrime%2Fcrime%2F01_District_wise_crimes_committed_IPC_2001_2012.csv?GoogleAccessId=gcp-kaggle-com@kaggle-161607.iam.gserviceaccount.com&Expires=1599078097&Signature=JDSu2lsgzbxZ4cxzIe4Ib4Tm6qYbCR4DpiywJPr5tbFQyTGGv0NO19slzn%2BWP0qDORJuKwhVhjvdo3J5%2FcPfjTWnRMKLD%2FHcnlksjbB%2F9m7LE0%2FGE%2FeQt4pBmaBaVVdT6%2F944FheXw0GXFE%2BomBRJOvme9hZOxQ2rmCxj4WsxhvLEgcKvgtUKytHES%2Fl%2FMwbvtF7gYuwzmK%2BmJL9yaO%2Fqyk3vwyUjNvYo7lf%2FMpWM8YAAnyC1osbta3hMPwEsf01K27GgVhCzMOykQAB45vZUvVfRoWU3o1%2BxvHKw8kdKVCL9ZSy7tefz8IB5nPydgJv8%2Bznfqun%2BudngkGe5C%2BPZA%3D%3D')
In [94]:
df = indcri
indcri.head()
Out[94]:
In [95]:
mumcri = df.loc[df['DISTRICT'] == 'MUMBAI']
mumcri.head(10)
Out[95]:
In [96]:
mumcri.columns
Out[96]:
In [97]:
plt.plot('YEAR', 'TOTAL IPC CRIMES', data=mumcri);
PyWaffle¶
In [98]:
import matplotlib as mpl
import matplotlib.pyplot as plt
import matplotlib.patches as mpatches
from pywaffle import Waffle
%matplotlib inline
In [99]:
scan = canada.loc[['Sweden', 'Denmark', 'Norway'], :]
scan
Out[99]:
In [100]:
cat = list(scan.index.values)
vals = scan['Total']
fig = plt.figure(FigureClass = Waffle, rows = 7, columns = 30, values = vals, labels = list(cat),
colors=["#2196f3", "#ff5252", "#999999"], figsize=(10,5),
legend = {'loc': 'lower left', 'bbox_to_anchor': (0.87, -0.4)})
In [101]:
indpakchi = canada.loc[['India', 'Pakistan', 'China'], :]
cat = list(indpakchi.index.values)
vals = indpakchi['Total']
fig = plt.figure(FigureClass = Waffle, rows = 7, columns = 30, values = vals, labels = list(cat),
colors=["#2196f3", "#ff5252", "#999999"], figsize=(10,5),
legend = {'loc': 'lower left', 'bbox_to_anchor': (0.878, -0.4)})
